用四元数与哈达玛变换抑制对抗噪声，提升恶劣天气图像恢复模型的鲁棒性。
Quaternion-Hadamard Network: A Novel Defense Against Adversarial Attacks with a New Dataset
- 在四元数域和变换域联合设计去噪模块，抑制高频对抗扰动。
- 在雨、雪、雾去除任务中，相比现有方法提升恢复质量与抗攻击能力。
- 适合需要高鲁棒性的低层视觉任务，如自动驾驶图像处理。
恶劣天气图像恢复（如雨、雪、雾）模型对基于梯度的白盒对抗攻击仍极为脆弱，微小的损失对齐扰动即可导致恢复输出严重退化。本文提出QHNet，一种计算高效的净化型防御方法，部署于恢复网络之前，通过在变换域与四元数域同时抑制扰动。QHNet在编码器-解码器框架中引入四元数哈达玛多项式去噪块（QHPDB）与四元数去噪残差块（QDRB），在保留精细结构细节的同时消除高频对抗噪声。使用PSNR与SSIM评估其在雨、雪、雾去除任务中的鲁棒性，并在采用投影梯度下降（PGD）、反向传播可微分近似（BPDA）及期望变换（EOT）的自适应、防御感知白盒攻击下验证。实验表明，相较于当前最优净化基线，QHNet在恢复保真度与鲁棒性上均有显著提升，证实其在低层视觉流水线中的有效性。
原文摘要 · Abstract (English)
Adverse-weather image restoration (e.g., rain, snow, haze) models remain highly vulnerable to gradient-based white-box adversarial attacks, wherein minimal loss-aligned perturbations cause substantial degradation in the restored output. This paper presents QHNet, a computationally efficient purification-based defense that precedes the restoration network and targets perturbation suppression in the transform and quaternion domains. QHNet incorporates a Quaternion Hadamard Polynomial Denoising Block (QHPDB) and a Quaternion Denoising Residual Block (QDRB) within an encoder-decoder framework to remove high-frequency adversarial noise while preserving fine structural details. Robustness is evaluated using PSNR and SSIM across rain, snow, and haze removal tasks, and further validated under adaptive, defense-aware white-box attacks employing Projected Gradient Descent (PGD), Backward Pass Differentiable Approximation (BPDA), and Expectation Over Transformation (EOT). Experimental results demonstrate that QHNet delivers superior restoration fidelity and significantly improved robustness compared to state-of-the-art purification baselines, confirming its effectiveness for low-level vision pipelines.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。